Configuring continuity

Mode, take-over, retention, history and visibility: what the individual settings do and when which choice makes sense.

The settings sit under continuity in the settings area. They apply to the entire tenant.

Activate continuity

The main switch. If it is off, no continuity copies are created and the view stays empty.

Mode

Automatic is the default and the normal case. Conbool recognises the outage itself: after three consecutive unsuccessful delivery attempts to the same domain, provided the first of them is at least five minutes old. The time condition prevents a brief hiccup from counting as an outage. Two successful deliveries switch back.

Permanently on is meant for planned maintenance. Even the first failed attempt leads to storage, without waiting for three failures. If delivery runs normally, nothing is created nonetheless. Anyone planning a mail server change switches this mode on beforehand and back afterwards.

This mode is at the same time the simplest way to try continuity once without waiting for a real disruption.

Suspended halts continuity without switching the module off. Useful when a known problem at the target system would otherwise constantly count as an outage.

Take-over: accept and deliver later

That is level 2. Without it, Conbool continues to answer the delivering server with a temporary rejection, and the message exists twice: there and as a reading copy here. With it, Conbool accepts the message and is solely responsible for it from then on.

Take-over requires a synchronised user directory. Without a directory it cannot be checked whether a recipient exists at all, and Conbool would accept mail for invented addresses that nobody ever collects.

In automatic mode take-over only applies once the outage is recognised. In permanently-on mode it applies from the first failed attempt.

Retention of the continuity copy

Between one and ninety days. After expiry it is deleted, regardless of whether the message was delivered.

This is outage support and not an archive. A long period brings little here: after the catch-up delivery the message sits in the mailbox anyway and, if booked, in the archive. It mainly creates a second place where mail lies.

Recording the history

So that during an outage the earlier thread is readable too and not only the backed-up mail. Two sources are available.

Mail preview uses the same storage as the preview in the delivery log. No second copy is created. The period follows the retention above and is at most fourteen days. Incoming mail only.

Archive reaches further back and is available as soon as the archive module is booked.

Visibility for members

Concerns only the navigation entry for members without administrative rights. Whether copies are created does not depend on it, and anyone who may configure continuity always sees the entry.

Always shows it permanently. Only in an emergency shows it as soon as an outage is recognised or readable copies are still held. Hidden never shows it, copies continue to be created but are not retrievable for members.

Visibility and mode are deliberately separate. Otherwise a maintenance window would inevitably be shown to all members.

Emergency contact address

An additional address for the outage notification, sensibly outside your own domains. The reasoning is described under preparation.

Access for the emergency

Not a switch but a status display: how many mailboxes have an account and how many members can sign in independently of the mail system. From here all missing accounts can be invited at once.

This is the most important section of the page. Everything else decides how well continuity works; this one decides whether anyone gets to see the result at all.